Image-Text-to-Text
Transformers
Safetensors
qwen3_vl
qwen3-vl
viss
streaming-video
visual-question-answering
causal-video
conversational
Instructions to use mis2/ViSS-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mis2/ViSS-8B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="mis2/ViSS-8B") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("mis2/ViSS-8B") model = AutoModelForMultimodalLM.from_pretrained("mis2/ViSS-8B", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use mis2/ViSS-8B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "mis2/ViSS-8B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mis2/ViSS-8B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/mis2/ViSS-8B
- SGLang
How to use mis2/ViSS-8B with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "mis2/ViSS-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mis2/ViSS-8B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "mis2/ViSS-8B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mis2/ViSS-8B", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use mis2/ViSS-8B with Docker Model Runner:
docker model run hf.co/mis2/ViSS-8B
Download code/viss_core/evaluation_protocol.py from mis2/ViSS-8B: direct link, hf CLI and curl.
- Browser
- Download file 5.65 kB
-
https://huggingface.co/mis2/ViSS-8B/resolve/main/code/viss_core/evaluation_protocol.py
- Command line
-
hf download hf://mis2/ViSS-8B/code/viss_core/evaluation_protocol.py
-
curl -L -o evaluation_protocol.py https://huggingface.co/mis2/ViSS-8B/resolve/main/code/viss_core/evaluation_protocol.py
5.65 kB
| from __future__ import annotations | |
| import hashlib | |
| import json | |
| import re | |
| from collections import Counter | |
| from pathlib import Path | |
| from typing import Any | |
| PROJECT_ROOT = Path(__file__).resolve().parents[1] | |
| LOCKED_PROTOCOL = PROJECT_ROOT / "configs/evaluation/recent4_full_v1.json" | |
| class ProtocolValidationError(ValueError): | |
| """Raised when the locked evaluation protocol or its exports drift.""" | |
| def load_locked_protocol(path: Path = LOCKED_PROTOCOL) -> dict[str, Any]: | |
| protocol = json.loads(path.read_text(encoding="utf-8")) | |
| if protocol.get("status") != "locked": | |
| raise ProtocolValidationError(f"{path}: protocol is not locked") | |
| if protocol.get("protocol_id") != "recent4_full_v1": | |
| raise ProtocolValidationError(f"{path}: unexpected protocol_id") | |
| inference = protocol.get("inference") or {} | |
| required_inference = { | |
| "causal_prefix_only": True, | |
| "sampling_fps": 1.0, | |
| "context_policy": "recent_n", | |
| "recent_frames": 4, | |
| "allow_future_frames": False, | |
| "allow_global_frames": False, | |
| "allow_extra_frame_inspection": False, | |
| "allow_persistent_visual_or_text_memory": False, | |
| } | |
| for key, expected in required_inference.items(): | |
| if inference.get(key) != expected: | |
| raise ProtocolValidationError( | |
| f"{path}: inference.{key}={inference.get(key)!r}, expected {expected!r}" | |
| ) | |
| processor = inference.get("processor_contract") or {} | |
| if not re.fullmatch(r"[0-9a-f]{64}", str(processor.get("preprocessor_config_sha256", ""))): | |
| raise ProtocolValidationError("Configure the protocol with your processor identity first") | |
| required_processor = { | |
| "image_processor_type": "Qwen2VLImageProcessorFast", | |
| "patch_size": 16, | |
| "merge_size": 2, | |
| "temporal_patch_size": 2, | |
| "shortest_edge": 65536, | |
| "longest_edge": 16777216, | |
| "reference_visual_tokens_per_frame": 576, | |
| } | |
| for key, expected in required_processor.items(): | |
| if processor.get(key) != expected: | |
| raise ProtocolValidationError( | |
| f"{path}: inference.processor_contract.{key}=" | |
| f"{processor.get(key)!r}, expected {expected!r}" | |
| ) | |
| return protocol | |
| def validate_locked_protocol(path: Path = LOCKED_PROTOCOL) -> dict[str, Any]: | |
| protocol = load_locked_protocol(path) | |
| root = path.resolve().parents[2] | |
| reports: dict[str, Any] = {} | |
| for benchmark, benchmark_config in protocol["benchmarks"].items(): | |
| export = root / benchmark_config["export"] | |
| rows = _load_jsonl(export) | |
| expected_count = int(benchmark_config["expected_queries"]) | |
| if len(rows) != expected_count: | |
| raise ProtocolValidationError( | |
| f"{export}: {len(rows)} rows, expected {expected_count}" | |
| ) | |
| digest = _sha256(export) | |
| if digest != benchmark_config["export_sha256"]: | |
| raise ProtocolValidationError( | |
| f"{export}: sha256 {digest}, expected {benchmark_config['export_sha256']}" | |
| ) | |
| uids = [str(row.get("uid") or "") for row in rows] | |
| if any(not uid for uid in uids) or len(set(uids)) != len(uids): | |
| raise ProtocolValidationError(f"{export}: missing or duplicate uid") | |
| labels = Counter(str(row.get("benchmark") or "") for row in rows) | |
| if labels != Counter({benchmark: expected_count}): | |
| raise ProtocolValidationError(f"{export}: benchmark labels {dict(labels)}") | |
| report: dict[str, Any] = { | |
| "export": str(export.relative_to(root)), | |
| "sha256": digest, | |
| "queries": len(rows), | |
| } | |
| expected_groups = benchmark_config.get("expected_groups") | |
| if expected_groups is not None: | |
| actual_groups = Counter( | |
| str((row.get("metadata") or {}).get("group") or "") for row in rows | |
| ) | |
| if dict(actual_groups) != expected_groups: | |
| raise ProtocolValidationError( | |
| f"{export}: groups {dict(actual_groups)}, expected {expected_groups}" | |
| ) | |
| report["groups"] = dict(sorted(actual_groups.items())) | |
| expected_tasks = benchmark_config.get("expected_tasks") | |
| if expected_tasks is not None: | |
| actual_tasks = Counter( | |
| str((row.get("metadata") or {}).get("task") or "") for row in rows | |
| ) | |
| if dict(sorted(actual_tasks.items())) != expected_tasks: | |
| raise ProtocolValidationError( | |
| f"{export}: tasks {dict(actual_tasks)}, expected {expected_tasks}" | |
| ) | |
| report["tasks"] = dict(sorted(actual_tasks.items())) | |
| reports[benchmark] = report | |
| return { | |
| "status": "locked_protocol_valid", | |
| "protocol_id": protocol["protocol_id"], | |
| "protocol_path": str(path.resolve().relative_to(root)), | |
| "benchmarks": reports, | |
| } | |
| def _load_jsonl(path: Path) -> list[dict[str, Any]]: | |
| rows = [] | |
| with path.open(encoding="utf-8") as handle: | |
| for line_number, line in enumerate(handle, start=1): | |
| if not line.strip(): | |
| continue | |
| value = json.loads(line) | |
| if not isinstance(value, dict): | |
| raise ProtocolValidationError(f"{path}:{line_number}: expected object") | |
| rows.append(value) | |
| return rows | |
| def _sha256(path: Path) -> str: | |
| digest = hashlib.sha256() | |
| with path.open("rb") as handle: | |
| for block in iter(lambda: handle.read(8 << 20), b""): | |
| digest.update(block) | |
| return digest.hexdigest() | |